We examined the site of action of nutrients that enhance mucosal permeability by use of D-glucose as an archetype of nutrients of this class. We tested the hypothesis that D-glucose enhances mucosal permeability by either acting outside the intestinal lumen after absorption or acting inside the intestinal lumen to cause mediator release from either endocrine or nerve tissue. The rate of absorption of L-[14C]glucose, a passively absorbed molecule, from the lumen was used as an index of the permeability of the mucosa of a perfused segment. L-Glucose was absorbed more rapidly in the presence of D-glucose than in the presence of an equimolar concentration of mannitol. However, the permeability of the jejunal mucosa was unaffected by elevated blood glucose levels during intravenous infusion of D-glucose. The mucosal permeability was also unaffected by exposure of an adjacent segment to D-glucose, a result suggesting that D-glucose does not alter mucosal permeability by inducing the release of a blood-borne mediator from enteroendocrine cells. Finally, the effect of D-glucose on mucosal permeability could not be blocked by hexamethonium or tetrodotoxin, a result suggesting that the intestinal nerves do not mediate this phenomenon. Lidocaine significantly increased the rate of L-glucose absorption when D-glucose was present in the lumen but had no effect on L-glucose absorption under basal conditions. Our findings indicate that D-glucose must be in contact with the apical membranes of enterocytes to alter mucosal permeability. This suggestion is consistent with the hypothesis that the modulation of mucosal permeability results from the activation of sodium-dependent cotransport systems.